Decadal Morphological Evolution and Governance Measures of the South Branch, Changjiang Estuary

Abstract

Estuaries and deltas hold significant socioeconomic importance and immense ecological value due to their dynamic geomorphic processes and unique geographical advantages. However, in recent decades, delta recession and the instability of river regimes have become global challenges, driven by intensive human interventions in upstream river basins and local regions. This study examines the South Branch of the Changjiang Estuary as a typical case to investigate its morphological evolution over the past decades and project future trends, offering suitable solutions to enhance the river regime stability. Analysis of bathymetric data reveals substantial channel–shoal adjustments in the South Branch from 1958 to 2016, characterized by significant erosion and deposition on a decadal scale. After 1997, reduced fluvial sediment supply has led to widespread erosion in the South Branch. Further disturbances at the Baimao Shoal and Biandan Shoal have exacerbated the instability of the river regime. Numerical predictions indicate continued erosion in the South Branch over the next 20 years, accompanied by further channel–shoal pattern adjustments. Hydrodynamic modeling of proposed measures demonstrates an improved flow ratio for the North Baimao Shoal Channel, contributing to enhanced channel–shoal system stability. These integrated governance measures have been incorporated into the latest renovation plan for the Changjiang Estuary. The findings provide valuable scientific guidance for the comprehensive management of the Changjiang Estuary and offer insights applicable to other large estuaries facing similar challenges.

1. Introduction

Estuaries serve as transitional zones connecting rivers and oceans, where terrestrial sediment is deposited and transported, shaping their complex and diverse environments and geomorphic forms [1]. As one of the most dynamic interfaces of land–sea interaction, estuaries and deltas offer abundant natural resources, including land, freshwater, fossil energy, agriculture, navigation, and ecological services. Consequently, many of the world’s large estuaries and deltas are densely populated and economically developed [2]. The unique geographical advantages of estuaries grant them significant socioeconomic and ecological value [3]. However, global climate change and human activities have introduced unprecedented challenges to their morphological evolution and sustainable utilization. Many estuaries and deltas worldwide are transitioning from accretion to erosion, a shift largely driven by reduced river sediment supply [2]. Notable examples include the Nile, Mississippi, and Mekong deltas. Similarly, China’s three largest estuaries—the Yellow, Changjiang, and Pearl—are all experiencing varying degrees of erosion, rendering them high-risk areas [4]. Understanding the morphological processes and long-term trends of estuaries on a decadal timescale and devising sustainable management strategies have become critical priorities for the academic community and policymakers alike.

The morphodynamics of estuaries are primarily governed by the interactions of river flow, tidal forces, and waves [5]. However, natural evolutionary processes have been disrupted by changes in river inputs, while local human interventions have further complicated estuarine responses [6,7,8]. The impacts of reduced fluvial sediment supply and human activities on estuarine morphology and regime stability have been extensively studied through data-driven analyses and process-based modeling. Data-driven analysis enhances our understanding of the characteristics and drivers of morphological changes using topographic and bathymetric datasets, and has been successfully applied in numerous case studies [6,8,9,10,11,12,13]. Process-based modeling, on the other hand, clarifies the intrinsic mechanisms linking physical forcing to morphological responses, facilitating predictions of future trends in estuarine evolution [14,15]. Techniques such as morphology acceleration allow for medium- and long-term trend predictions [16]. Integrating these two approaches is particularly effective for uncovering evolutionary mechanisms, forecasting future morphological changes, and developing robust governance strategies.

The Changjiang Estuary, characterized by its large-scale and three-level bifurcated structure, is the outlet of the largest and longest river in Asia (Figure 1a). Since the 1980s, human activities have significantly influenced river discharge and sediment load in the catchment, while major engineering projects within the estuary have been undertaken since the late 1990s. Reduced sediment supply has shifted the estuary from a state of net accretion to net erosion, affecting areas from the inner estuary to the mouth bar and subaqueous delta front [6,17,18]. Estuarine engineering projects have also contributed to localized morphological changes, with varying impacts observed across different bifurcation channels and shoals [8,19,20,21,22]. As the primary bifurcation of the Changjiang Estuary, the South Branch is the main conduit for freshwater discharge, accounting for over 95% of runoff to the sea. It is home to numerous harbors, infrastructure, and freshwater reservoirs, and serves as the sole 12.5-meter-deep navigation channel linking the Changjiang Golden Waterway to international shipping. Thus, maintaining the stability of the South Branch’s river regime is crucial for regional economic development. Previous studies have investigated the recent morphological changes in the South Branch and identified controlling factors [23,24,25]. Continuous erosion of the riverbed has been observed since the closure of the Three Gorges Dam in 2003, alongside self-adaptive adjustments in the river regime to changing conditions [26]. Additionally, the flow diversion pattern around the Baimao Shoal has evolved, while the Biandan Shoal remains the largest uncontrolled shallow shoal in the inner estuary. These factors contribute to the instability of the river regime, highlighting the urgent need for systematic research on governance measures—an imperative that forms the primary motivation of this study.

The aims of this paper are to (1) explore the stability of the river regime in the South Branch using historical bathymetric datasets and numerical predictions, and (2) propose and evaluate governance measures to stabilize the river regime. These research findings not only provide guidance for the comprehensive management of the Changjiang Estuary but also support high-quality regional social and economic development. The rest of the article is structured as follows. Section 2 introduces the study area. Section 3 describes the data and methods used in the study. Section 4 presents and discusses the numerical results. Finally, conclusions are provided in Section 5.

2. Study Area

The Changjiang Estuary is shaped by abundant river runoff and sediment load, influenced by river flow, astronomical tides, and wind waves. This dynamic system has formed a distinct landform characterized by a three-level bifurcation and a four-outlet configuration [27]. Spanning from Xuliujing to the outer subaqueous delta, the Changjiang Estuary extends over 180 km. The channel width expands from approximately 5 km in the inner estuary to over 90 km in the mouth bar area (Figure 1b). Within the Changjiang Estuary, multiple channel–shoal systems are present, varying in size from several kilometers to dozens of kilometers. River runoff has remained relatively stable since the 1950s, with occasional fluctuations during particularly wet or dry years. However, sediment load began to decline in the 1980s due to river damming, and soil and water conservation efforts in the basin. This trend continued, reaching a significantly reduced level of 134 Mt/a following the closure of the Three Gorges Dam in 2003 (Figure 2). The mean and maximum tidal range values in the Changjiang Estuary are 2.66 m and 4.62 m, respectively [28]). As a result of the interplay between high river discharge and moderate tidal currents, the Changjiang Estuary is classified as a joint river- and tide-dominated estuary.

The South Branch represents the first-level bifurcation of the Changjiang Estuary, extending from Xuliujing to Wusong, with a length of approximately 70.5 km and a width ranging from 5.7 to 15 km. The main channel is divided into two secondary channels by the Baimao Shoal, whose head was stabilized by engineering works in 2014 to improve navigation. The Biandan Shoal stretches from the south bank of Chongming Island to the terminus of the South Branch, forming a large-scale channel–shoal system alongside the flood-dominated Xinqiao Channel and the ebb-dominated main channel. River regulation projects, including the Qingcaosha Reservoir, the Xinliuhe Shoal protection works, and the submerged breakwater at the Nanshatou Channel that was completed in 2009, have permanently fixed the bifurcation point between the North and South Channels. Apart from the Biandan Shoal, the overall channel–shoal configuration within the South Branch has remained relatively stable with the implementation of these projects.

3. Data and Methods

3.1. Bathymetric Data

To quantify decadal morphological changes in the South Branch, we analyzed bathymetry data collected over multiple years (1958, 1978, 1986, 1997, 2007, and 2016). The bathymetry data for 1958, 1978, and 1986 were digitized from marine charts published by the Navigation Guarantee Department of the Chinese Navy Headquarters, while data for subsequent years were graciously provided by the Changjiang Estuary Waterway Administration Bureau of the Ministry of Transport. Most measurements were conducted following periods of river flooding. Water depths were measured using an echo sounder, and positions were recorded using a Global Positioning System (GPS). The bathymetric maps were scaled between 1:25,000 and 1:50,000, ensuring sufficient depth point density for accurately calculating erosion and accretion volumes [13]. A Digital Elevation Model (DEM) was created using Geographic Information System (GIS) software , with a grid resolution of 50 × 50 m generated via the Kriging interpolation method. Erosion and deposition patterns were derived by subtracting DEM grids from later years against those from earlier years. The domain for calculating channel volume and erosion/accretion volumes is illustrated in Figure 1b.

3.2. Numerical Model

3.2.1. Governing Equations

In this study, the hydrodynamic module and mud transport module of the MIKE 21 flow model [29] are utilized jointly for numerical simulations. The hydrodynamic model is a depth-averaged, two-dimensional framework based on unstructured grids. The governing equations of the two-dimensional numerical model include the continuity equation for water flow, the momentum equation for water flow, the transport equation for viscous sediment, and the bed surface deformation equation, which are expressed as follows:

(1)

continuity equation of water flow:

$${\displaystyle \frac{\partial Z}{\partial t}}+{\displaystyle \frac{\partial uH}{\partial x}}+{\displaystyle \frac{\partial vH}{\partial y}}=q.$$

(1)

(2)

momentum equation of water flow:

$$\begin{array}{cc}\hfill {\displaystyle \frac{\partial uH}{\partial t}}& +{\displaystyle \frac{\partial uuH}{\partial x}}+{\displaystyle \frac{\partial vuH}{\partial y}}=-g{\displaystyle \frac{n^2\sqrt{u^2+v^2}}{H^{1/3}}}u-gH{\displaystyle \frac{\partial Z}{\partial x}}\hfill \\ \hfill & +{\displaystyle \frac{\partial }{\partial x}}\left(v_T{\displaystyle \frac{\partial uH}{\partial x}}\right)+{\displaystyle \frac{\partial }{\partial y}}\left(v_T{\displaystyle \frac{\partial uH}{\partial y}}\right)+q{u}_0+f_{0}Hv,\hfill \end{array}$$

(2)

$$\begin{array}{cc}\hfill {\displaystyle \frac{\partial vH}{\partial t}}& +{\displaystyle \frac{\partial uvH}{\partial x}}+{\displaystyle \frac{\partial vvH}{\partial y}}=-g{\displaystyle \frac{n^2\sqrt{u^2+v^2}}{H^{1/3}}}v-gH{\displaystyle \frac{\partial Z}{\partial y}}\hfill \\ \hfill & +{\displaystyle \frac{\partial }{\partial x}}\left(v_T{\displaystyle \frac{\partial vH}{\partial x}}\right)+{\displaystyle \frac{\partial }{\partial y}}\left(v_T{\displaystyle \frac{\partial vH}{\partial y}}\right)+q{v}_0-f_{0}Hu.\hfill \end{array}$$

(3)

(3)

transport equation of viscous sediment:

$${\displaystyle \frac{\partial C}{\partial t}}+u{\displaystyle \frac{\partial C}{\partial x}}+v{\displaystyle \frac{\partial C}{\partial y}}={\displaystyle \frac{1}{H}}{\displaystyle \frac{\partial }{\partial x}}\left(H{D}_x{\displaystyle \frac{\partial C}{\partial x}}\right)+{\displaystyle \frac{1}{H}}{\displaystyle \frac{\partial }{\partial y}}\left(H{D}_x{\displaystyle \frac{\partial C}{\partial y}}\right)+q{C}_L{\displaystyle \frac{1}{H}}+{\displaystyle \frac{E-D}{H}}.$$

(4)

(4)

deformation equation of the bed surface:

$${\gamma }_b{\displaystyle \frac{\partial z_b}{\partial t}}=\sum (D-E),$$

(5)

$$D={\omega }_s{c}_b{p}_d,$$

(6)

$$E=\epsilon {\left({\displaystyle \frac{{\tau }_b}{{\tau }_{ce}}}-1\right)}^s.$$

(7)

where Z is the water level; H is the water depth; u and v are the average velocity of water depth in the x and y directions, respectively; q is the source sink strength of water flow per unit area; n is the roughness; g is gravitational acceleration; $v_T$ is the comprehensive diffusion coefficient of water flow, which is generally equivalent to the turbulent viscosity coefficient; $u_0$ and $v_0$ are the components of the momentum source (sink) in the x and y directions, respectively; $f_0$ is the velocity coefficient of Coriolis force; C is the sediment content; $D_x$ is the diffusion coefficient of suspended sediment; $C_L$ is the sediment concentration of the source; D and E represent the settlement and suspended sediment flux per unit area of the bed surface, respectively; $z_b$ is the elevation of the riverbed; ${\gamma }_b$ is the dry density of riverbed sediment; ${\omega }_s$ is the sedimentation velocity of sediment; $c_b$ is the concentration of suspended sediment near the bottom; $p_d$ is the probability of settlement; $\epsilon$ is the scouring coefficient of the bed surface; ${\tau }_b$ is the shear stress of the bed surface; ${\tau }_{ce}$ is the critical scouring shear stress of the bed surface; s is the scouring index; $p_d=1-{\displaystyle \frac{{\tau }_b}{{\tau }_{cd}}}$; and ${\tau }_{cd}$ is the critical settlement shear stress of the bed surface.

3.2.2. Set-Up of the Numerical Model

The numerical simulation domain (Figure 3) extends from Jiangyin (the tidal current limit) to the East China Sea, approximately 200–240 km beyond the mouth of the Changjiang River. Due to the large spatial scale of the model domain, varying grid resolutions were applied across different subareas. The roughness coefficient was determined empirically and exhibited a decreasing trend from upstream to downstream. Grid sizes ranged from approximately 200 m in the upper reaches of the Changjiang Estuary to about 600 m in the mouth area, progressively increasing to around 10 km in the offshore region. The total number of grid cells was 71,778. The roughness coefficient distribution within the model domain followed the values adopted by [30]. From Jiangyin to Xuliujing: decreasing from 0.021 to 0.015 ${\mathrm{m}}^{1/3}{\mathrm{s}}^{-1}$. In the North Branch: 0.010∼0.011 ${\mathrm{m}}^{1/3}{\mathrm{s}}^{-1}$. In the South Branch and other downstream river channels: 0.012∼0.014 ${\mathrm{m}}^{1/3}{\mathrm{s}}^{-1}$. Offshore outside the mouth: 0.010∼0.011 ${\mathrm{m}}^{1/3}{\mathrm{s}}^{-1}$. Deep sea area: 0.009∼0.010 ${\mathrm{m}}^{1/3}{\mathrm{s}}^{-1}$.

The open-sea boundary tidal levels were derived from the MIKE global tide model, with adjustments for local time zones and height data.

Sediment concentration at the open-sea boundary was set to zero. Testing a zero-gradient condition for this boundary revealed no significant differences in most areas. Water flow discharge and sediment concentration series were prescribed at Jiangyin, representing the upstream boundary of the Changjiang River.

Field measurements in August 2017 provided suspended sediment and bed load data for the Changjiang Estuary. Median diameters during spring and neap tides were 0.013/0.11 mm and 0.008/0.13 mm, respectively. Based on these data, two sediment types were simulated [31]:

(1) Viscous fine-grained sediment: particle size range of 0∼0.031 mm, with a representative size of 0.008 mm, incorporating flocculation effects. (2) Non-viscous fine-grained sediment: particle size range of 0.031∼1.0 mm, with a representative size of 0.11 mm, excluding flocculation. Considering the influence of salinity on sediment flocculation in the Changjiang Estuary, and based on prior research, the settling velocity of viscous sediment increases gradually from 0.1 mm/s at Jiangyin to 2.0 mm/s in the estuary. For non-viscous sediment, the settling velocity is 10.2 mm/s.

The riverbed thickness was set to 15 m to account for potential erosion. The proportions of viscous and non-viscous sediments in the riverbed were based on measured data, with viscous sediment comprising 10% and non-viscous sediment 90%. Key parameters, including the critical shear stress and erosion coefficient of the bed material, were calibrated and determined to fall within the following ranges: critical shear stress: 0.17∼0.27 $\mathrm{N}/{\mathrm{m}}^2$; and erosion coefficient: $1.5\times {10}^{-4}$ to $2.8\times {10}^{-4}$$\mathrm{kg}/\left({\mathrm{m}}^2\mathrm{s}\right)$. The recommended bed surface roughness height is 0.001 m.

3.2.3. Validation of the Numerical Model

The Changjiang River Estuary Bureau of Hydrology and Water Resources Survey conducted a comprehensive hydrological test covering a complete spring and neap tidal cycle during the flood season from 23 to 31 August 2017. For the purposes of this study, tidal levels and velocities at representative stations during the August 2017 flood season were used to validate the hydrodynamic parameters of the numerical model. The stations at Xuliujing and Chongtou were selected for tidal level validation. Xuliujing is located upstream of the bifurcation of the North and South Branches, while Chongtou is situated at the entrance of the North Branch. Additionally, sections BMS1 and BMS2 were chosen to validate tidal velocities. These sections are located between Baimaosha and Chongming Island, and between Baimaosha and the southern bank, respectively. The locations of the hydrological stations and monitoring sections are shown in Figure 4.

Figure 5 compares the simulated and measured tidal levels at Xuliujing and Chongtou. The results demonstrate that the phase and amplitude of the calculated tidal levels align well with the observed data. For tidal velocity validation, the calculated results at sections BMS1 and BMS2 were compared with measured data. Notably, the distances 900, 1400, 1800, and 2400 (in meters) represent the positions of the measured vertical lines from the left bank starting point of each cross-section. The comparison between simulated and measured flow velocities is illustrated in Figure 6. It can be seen from Figure 6 that the calculated flow velocity is in good agreement with the measured flow velocity.

To further calibrate the parameters used in the Mud Transport (MT) module, the morphological evolution from November 2011 to October 2016 was simulated. To reduce computational costs, the series of water flow discharge and sediment concentration at Jiangyin were averaged in two steps. First, a multi-year average condensed the five-year process into one year. Second, the year was divided into five periods, each containing five spring-neap tidal cycles (approximately 14.6 days per cycle), and an average was calculated for each period. These two averaging steps resulted in an evolution acceleration factor of 25. The upstream boundary conditions for water flow and sediment concentration were derived from a one-dimensional water flow and sediment transport model, covering the area from Datong to Xuliujing. Figure 7a,b present the observed and simulated riverbed evolution from November 2011 to October 2016. The results show that the simulated patterns of erosion and deposition align well with the observed data. However, calculation errors in the North and South Branches were larger than those in other sections, likely due to both model limitations and channel improvement activities, such as dredging, sand throwing, and the construction of submerged breakwaters. For example, measured sedimentation in the lower North Branch, especially near the edge of Chongming Island, was higher than simulated values. This discrepancy may be partly attributed to insufficient consideration of human activities and the impacts of vegetation on Chongming Island in the model. Overall, the validation results demonstrate the effectiveness and applicability of the numerical model. Further details on model setup and quantitative validation can be found in our previous study [32]

3.2.4. Model Setting for Evolution Prediction

The initial bathymetry used for the evolution prediction of the Changjiang Estuary was derived from topographic data surveyed in October 2016. Data on water and sediment discharge at Jiangyin were obtained from a one-dimensional water flow and sediment transport model, which encompasses the river reach from Datong to Xuliujing in the lower Changjiang River. For the evolution prediction, inflow processes of water discharge and sediment concentration from 2008 to 2017 were employed.

To ensure consistency with the validation process while minimizing computational costs, the inflow conditions for 2008–2017 were averaged. Figure 8 presents the boundary conditions for water discharge and sediment concentration at Jiangyin during a computational period. Each computational period was divided into five intervals, each lasting approximately 14.6 days, corresponding to a spring-neap tidal cycle. With an evolution acceleration factor of 25, a single computational period represented five years of bed evolution. The evolution prediction consisted of four computational periods, collectively spanning 20 years of morphological changes.

3.2.5. Model Setting for Measure Comparison

To identify effective governance measures for stabilizing the river regime, we compared the effects of different measures using the validated model. Given that the uncertainty in simulating bed evolution was significantly higher than that in simulating water flow movement, we performed hydrodynamic simulations solely for the purpose of comparing the various measures. The comparison period lasted 24 h, from 00:00 on 23 September to 00:00 on 24 September 2002. During this period, the average inflow discharge from the Changjiang River at Jiangyin was approximately 36,000 m³/s, representing a medium flow, and the outer sea was a spring tide.

To assess the impact of the governance measures on tidal levels and flow velocities, monitoring points were established in each waterway and on shallow shoals. These points are shown in Figure 9 and listed in

Table 1. Names of monitoring points.

ID	Name	ID	Name	ID	Name
p1	Wangyu River	p11	The South Passage	p21	Lianxing Port
p2	Xuliujing	p12	Liuxiao	p22	Main Channel at Xuliujing
p3	Outlet of Baimao River	p13	Baozheng	p23	South Baimao Shoal Channel
p4	Outlet of Dangxi River	p14	Nanmen	p24	North Baimao Shoal Channel
p5	Qiyakou	p15	Chongtou	p25	Upper Biandan Shoal
p6	Outlet of Liu River	p16	Rixinhe	p26	Lower Biandan Shoal
p7	Outlet of Wusong River	p17	Qilong Port	p27	Main Channel of the South Branch
p8	Hengsha Shoal	p18	Lingdian Port	p28	Xinqiao Connecting Channel
p9	The North Passage	p19	Touxing Port	p29	Main Channel of the South Channel
p10	The North Channel	p20	Santiao Port	p30	Main Channel of the North Channel

. Additionally, cross-sections were set up in each channel, and the sectional flow discharge modeling results were used to calculate the diversion ratio. The calculation method for the dispersion ratio involves three steps. First, the ebb (or flood) volume of each forked river channel is determined for the same tidal cycle. Second, the total ebb (or flood) volume across all forked river channels is calculated for that cycle. Finally, the dispersion ratio for each channel is computed as the proportion of its ebb (or flood) volume to the total ebb (or flood) volume. For example, the ebb tide dispersion ratio in Nangang within the Changjiang Estuary can be expressed as:$\mathrm{Dispersion}\mathrm{Ratio}\left(\mathrm{Nangang}\right)={\displaystyle \frac{\mathrm{Ebb}\mathrm{Tide}\mathrm{Volume}\mathrm{in}\mathrm{Nangang}}{\mathrm{Ebb}\mathrm{Tide}\mathrm{Volume}\mathrm{in}\mathrm{Nangang}+\mathrm{Ebb}\mathrm{Tide}\mathrm{Volume}\mathrm{in}\mathrm{Beigang}}}.$ The locations of these sections are depicted in Figure 9.

4. Results and Discussion

4.1. Observed Morphological Changes

The South Branch underwent substantial channel–shoal adjustments between 1958 and 2016, characterized by notable temporal and spatial variations. The Baimao Shoal formed in the 1950s as fluvial sediment accumulated in the middle of the channel due to a sudden widening and a corresponding decrease in flow velocity downstream of the Xuliujing node (Figure 10a). Following its formation, the Biandan Shoal gradually expanded and increased in elevation, reaching its maximum size around 1997 (Figure 10b–d). Since then, the channel–shoal system has remained relatively stable, further reinforced by engineering projects at the head of the Baimao Shoal in 2014. Between 2002 and 2012, the ebb diversion ratio of the South Baimao Shoal Channel (the river channel on the south side of Baimao Shoal) increased from 57.2% to 72.7%, while the North Baimao Shoal Channel (on the north side) experienced silting and shrinkage. This deposition in the North Baimao Shoal Channel was largely attributed to extreme flood events in the Changjiang River and sediment flow reversal from the North Branch [33]. Downstream, the Baimao Shoal, Biandan Shoal, the ebb-dominant main channel, and the flood-dominant Xinqiao Channel form a large-scale compound channel–shoal system, a feature commonly observed in tidal rivers worldwide [34]. The Biandan Shoal, formed over a century, continues its natural evolution. The Upper Biandan Shoal, attached to Chongming Island in a relatively fixed position, contrasts with the Lower Biandan Shoal, which has migrated downstream and narrowed the Xinqiao Connecting Channel. Significant adjustments occurred at the end of the South Branch between the late 1970s and mid-1980s, when shoal incision and merging reshaped the channel–shoal pattern in a relatively short time (Figure 10b,c). After the late 1980s, the Lower Biandan Shoal entered a new phase of evolution [6]. The construction of training jetties at the head of the Xinliuhe Shoal in 2009 further weakened ebb diversion through the Xinqiao Connecting Channel, intensifying the incision of the Biandan Shoal (Figure 10e,f).

Erosion and deposition within the South Branch were significant between 1958 and 2016. From 1958 to 1978, deposition primarily occurred at the Baimao Shoal and the Lower Biandan Shoal due to shoal growth and downstream migration, respectively (Figure 11a). Between 1978 and 1986, the patterns of erosion and accretion near the end of the Biandan Shoal reversed compared with the earlier period (Figure 11b). This reversal was driven by the incision of the Biandan Shoal and the integration of separated shoals into downstream formations. After 1986, erosion became widespread across most areas of the South Branch, particularly during the period from 1986 to 1997 (Figure 11c). Previous studies attribute this shift to a decline in fluvial sediment supply during the 1980s and frequent river floods in the 1990s [6]. After 1997, similar patterns of erosion and deposition persisted, with overall erosion dominating the South Branch from 2007 to 2016 (Figure 11d,e).

To analyze these changes, a hypsometric curve (Figure 12) was generated by plotting water surface area data under various hypothetical water level conditions for a given river section. The slope changes of the curve reveal the vertical geometric characteristics of the river section, distinguishing wide-shallow profiles from narrow-deep ones. Comparing the curves across different years highlights trends in erosion and deposition at varying water depths. For the South Branch, the water surface area above −15 m initially decreased and then increased, while the water surface area below −15 m consistently diminished. The hypsometric curves reveal an inflection point between −10 m and −15 m, which separates the shoals from the main channel. These patterns indicate that the riverbed experienced accretion and subsequent erosion between 0 m and −15 m, while the deep channel below −15 m was continuously eroded and deepened. Consequently, the South Branch evolved into a narrower and deeper system. This phenomenon can be attributed not only to the reduction in sediment supply but also to the significant role of coastal embankment and port construction in driving these changes.

4.2. Instability of the River Regime

The river regime of the South Branch is influenced by both river–tidal interactions and local engineering interventions. Although the Baimao Shoal underwent periods of division and merging during the 1980s, the double-bifurcation pattern has remained stable in recent decades. To stabilize the thalweg line near Xuliujing, artificial nodes were constructed through riverbank protection measures to mitigate erosion. The artificial node on the north bank was established through the “Xintonghai Sand Enclosure Project”, while the node on the south bank resulted from the “Changshu Beach Enclosure Project”. Following the creation of these nodes, oscillations in the mainstream flow at Xuliujing were reduced. Additionally, the southward mainstream was deepened and narrowed due to riverbed scouring, which compensated for the reduced fluvial sediment discharge [25]. This process facilitated the rapid development of the South Baimao Shoal Channel. However, the resulting steepening of the south bank slope posed safety risks to Taicang Port and its embankments [35]. Furthermore, the development of the South Baimao Shoal Channel has primarily promoted water inflow to the North Channel rather than the South Channel. This ongoing trend could destabilize the river regime in the lower section of the South Branch and jeopardize the deep navigation channels in the South Channel and North Passage.

In the Biandan Shoal system, the mainstream of the Xinqiao Connecting Channel has migrated downstream and become narrower as the Lower Biandan Shoal has shifted. This migration has caused erosion near the upstream side of the Qingcaosha Reservoir, posing a threat to the safety of the reservoir embankment. Additionally, the narrowing of the Xinqiao Connecting Channel has reduced its diversion capacity, indirectly intensifying the ebb flow near the upstream side of the Biandan Shoal. The strengthened ebb flow over the Biandan Shoal has exacerbated channel scouring, particularly during river floods. Furthermore, the increased water level gradient between the north and south sides of the shoal has accelerated its incision, potentially triggering erosion along the Xinqiao Channel. These processes have destabilized the river regime of the South Branch, complicating efforts for the comprehensive management and utilization of the Changjiang Estuary.

4.3. Numerical Predictions for Future Evolution

The numerical predictions of cumulative deposition and erosion thickness in the South Branch over the next 20 years are shown in Figure 13. The results indicate that significant scouring is expected in the central region of the bifurcation area between the South and North Branch, with depths ranging from 2 to 10 m. In contrast, deposition of approximately 0 to 10 m is anticipated at the corner of the northern waterfront and the small shoal adjacent to the south bank. Additionally, the entrance to the North Branch is expected to become more curved, leading to a stronger influence of the rising tide in the North Branch. This phenomenon occurs because, as the entrance of the North Branch becomes more curved, it becomes more challenging for river runoff from upstream to enter the North Branch, thereby increasing the influence of tides entering through the entrance. The narrowing of the channels on both sides of the middle section of the North Branch will contribute to the further formation of a trumpet-shaped morphology in the middle and lower sections of the North Branch, eventually leading to the occurrence of tidal bores during high tide in the North Branch.

The Baimao Shoal is expected to experience significant siltation, while its neighboring troughs will undergo scouring. Specifically, the South Baimao Shoal Channel will experience a maximum scour thickness exceeding 10 m, while the North Baimao Shoal Channel will experience both mild erosion and mild deposition. As a result, the diversion ratio in the South Baimao Shoal Channel is expected to increase, which may negatively affect the stability of the diversion ratio in the South Channel.

The head of the Upper Biandan Shoal is expected to become silted, while its tail will undergo scouring, leading to an increased separation from the Lower Biandan Shoal. On the south side of the Lower Biandan Shoal, the Xinqiao Connected Channel will experience scouring. Additionally, sediment deposition will extend further at the tail of the Lower Biandan Shoal, with similar deposition occurring in the head area of the Xinliuhe Shoal, which connects to Changxing Island. The scoured zone between the Upper and Lower Biandan Shoals, particularly between Gelong Port and Nanmen Port, is referred to as the Xingelonggang Channel. The development of this scoured zone has facilitated the growth of the Xinqiao Channel, while simultaneously threatening the stability of the Lower Biandan Shoal. The head of the Lower Biandan Shoal is expected to undergo mild deposition, while the middle and lower sections of the shoal will experience more significant sediment deposition. The Xinqiao Channel on the northern side of the Lower Biandan Shoal will become silted, while the southern edge of the shoal will be subject to scouring, which will cause the Xinqiao Connecting Channel to widen.

Overall, the results suggest that the South Branch will experience net scouring, with the river regime remaining largely unchanged. However, without proper intervention, significant scouring and expansion are expected in both the South Baimao Shoal Channel and the Xinqiao Connecting Channel. At the same time, the North Baimao Shoal Channel and the entrance to the South Channel may decrease in size. This evolutionary trend could reduce the water flow diversion ratio in both the North Baimao Shoal Channel and the South Channel, which would be detrimental to maintaining a deep navigation channel in the South Channel. Additionally, the Upper and Lower Biandan Shoals are likely to become further separated, negatively impacting the stability of the Lower Biandan Shoal.

4.4. Effect of Governance Measures

4.4.1. Baimao Shoal

To identify the optimal governance measure for Baimao Shoal, we present four potential measures in

Table 2. Governance measures for the Baimao Shoal.

No.	Governance Measures for the Baimao Shoal
1	Lengthen the three tooth dikes at the south of the Baimao Shoal outward by 400 m, and keep the elevation of its head as −7 m.
2	Set a submerged dam in the South Baimao Shoal Channel with a crest elevation of −20 m.
3	Rebuild the thorn dam connecting to the west vertex of the Baimao Shoal, keep the connecting point still, and deflect it counterclockwise. The crest elevation of the new thorn dam is −3 m to 0 m, and the length is approximately 1.8 km.
4	Cover the Taicang beach in front of the right bank of the South Baimao Shoal Channel from Xinjing Gate to Lang Port. The covered area extends to a length of 13 km and has the lowest elevation of -20 m.

and Figure 14. These measures were all carefully screened in an earlier stage of the study. It is important to note that the recommended options are based on a quantitative comparison of the effects on the flow field under each option, with a focus on high and low tide levels at monitoring points and the tidal fluctuations at various monitoring sections. Due to the large number of monitoring points and sections involved, we have omitted the intermediate data in this article and will only provide a brief summary of the results here.

Each of the four measures was designed to reduce the diversion ratio in the South Baimao Shoal Channel. Among them, Option 2 proved to be the most effective, leading to a 20.84% increase in the diversion ratio of the ebb tide stream in the North Baimao Shoal Channel. In contrast, Options 1, 3, and 4 resulted in much smaller increases of 0.2%, 1.19%, and 1.46%, respectively.

Furthermore, the governance measures also affect the tide levels, which is a critical consideration for flood prevention and riverbank drainage. Option 2 causes a general rise in the highest tide levels at the monitoring stations, with the maximum increase ranging from approximately 0.059 to 0.068 m. In Options 3 and 4, the highest tide levels also increase, but the increments are less than 0.05 m. Under Option 1, the increments in the highest tide levels at the monitoring points are all less than 0.01 m. Based on these results, we conclude that Option 2 will result in higher water levels, while Option 1 will lead to lower water levels. Therefore, Options 3 and 4 are recommended.

4.4.2. Biandan Shoal

The stability of the Biandan Shoal is crucial for maintaining the overall stability of the river regime in the South Branch and the diversion between the South Channel and the North Channel. To stabilize the Biandan Shoal, a protective measure is necessary. In

Table 3. Governance measures for stabilizing Biandan Shoal.

No.	Content of the Measures
1	The content in the three options is the same and is as follows. The submerged breakwater at the right edge of the Upper Biandan Shoal is arranged along a −2 m contour line. The upper part of the submerged breakwater at the right edge of the Lower Biandan Shoal is along the −2 m contour line. The lower part is arranged along the −4 m contour line.	Crest elevations of the breakwaters are +2.5 m.
2		Crest elevations of the breakwaters are 0 m.
3		Crest elevations of the breakwaters are −0.5 m.

and Figure 14, we present three options, with the only difference being the crest level of the submerged breakwater.

Under the influence of the proposed measures, both the flood tide and ebb tide volumes in the South Branch decreased, with the flood tide volume experiencing a greater reduction than the ebb tide volume. Options 2 and 3 had minimal impact on the diversion ratio of the North and South Baimao Shoal Channels, while Option 1 slightly decreased the diversion ratio of the ebb tide stream in the North Baimao Shoal Channel. The diversion ratio between the South Channel and the North Channel showed more significant changes than those in the North and South Baimao Shoal Channels. Specifically, Options 1, 2, and 3 increased the diversion ratio of the ebb tide stream in the South Channel by 1.31%, 1.27%, and 0.84%, respectively. Given that the Changjiang Estuary deep-water channel passes through the South Channel, a moderate increase in the diversion ratio of the South Channel is considered beneficial for the stability of the system.

Considering the tidal levels, the effect of Option 1 was notably stronger than the other two options. All of the measures had only a slight impact on the high tidal level, with the highest tidal level decreasing in most areas. However, the low tide level in the South Branch was more significantly affected. The lowest tidal levels at monitoring points near the outlets of the Dangxi River and Liu River increased by approximately 0.050–0.100 m, 0.024–0.088 m, and 0.024–0.090 m under Options 1, 2, and 3, respectively. The rise in low tide levels presents challenges for riverbank drainage. Considering the overall impact of the measures, as well as factors such as project size and eco-friendliness, Option 3 is recommended for the governance of Biandan Shoal.

4.4.3. Combined Governance Measures

Governance measures have been planned for both Baimao Shoal and Biandan Shoal. The combined effects of multiple projects may differ from the simple superposition of the individual project impacts. To account for this, we tested the effectiveness of the combined governance measures, which are outlined in

Table 4. The combined measures for stabilizing the Baimao Shoal and Biandan Shoal.

No.	Baimao Shoal Governance Measures	Biandan Shoal Governance Measures
1	Rebuild the thorn dam connecting to the west vertex of the Baimao Shoal, keep the connecting point still, and deflect it counterclockwise. The crest elevation of the new thorn dam is −3 m to 0 m and the length is approximately 1.8 km.	The content in the four options is the same and is as follows. The submerged breakwater at the right edge of the Upper Biandan Shoal is arranged along the −2 m contour line. The upper part of the submerged breakwater at the right edge of the Lower Biandan Shoal is along the −2 m contour line. The lower part is arranged along the −4 m contour line. Crest elevations of the submerged breakwaters are −0.5 m.
2	Lengthen the three tooth dikes at the south of the Baimao Shoal outward by 400 m, and keep the elevation of its head as −7 m.
3	Set a submerged dam in the South Baimao Shoal Channel with a crest elevation of −20 m.
4	Measure content in both Options 1 and 2.

.

The simulation results indicated that, in general, the combined measures produced effects similar to those of the individual measures. Under Options 1, 2, and 4, the diversion ratios in the North Baimao Shoal Channel increased by approximately 0.4%, 0.5%, and 0.9%, respectively, for both the flood tide and ebb tide streams. However, Option 3 led to a significantly larger increase of 4–4.5% in the diversion ratio of the North Baimao Shoal Channel, which was much higher than the other three options. Regarding the diversion ratios in the North Channel, Option 3 showed a slightly weaker effect compared with the other three options. Under all four options, the diversion ratio of the ebb tide stream in the North Channel decreased, while the ratio of the flood tide stream increased. The changes in the diversion ratios in both the North and South Channels under Options 1, 2, and 4 were quite similar. These options resulted in a decrease of approximately 0.8% in the diversion ratio of the ebb tide stream and an increase of approximately 1.8% in the diversion ratio of the flood tide stream in the North Channel. In contrast, Option 3 caused smaller variations, with a 0.2% decrease in the ebb tide ratio and a 0.8% increase in the flood tide ratio.

Considering the high tide levels at the monitoring points, those located in the South Branch were the most affected under all four options. Option 3 had the most significant impact, with the highest tide levels increasing by more than 5 cm near Nanmen Port. In contrast, the increments in the other three options did not exceed 3 cm (for example, the variation in high tide levels under Option 4 is illustrated in Figure 15). The effect on the low tide levels under different options showed a similar trend. Option 3 led to increases in the lowest tide levels at points along the South Baimao Shoal Channel exceeding 10 cm, while the corresponding increments for the other three options did not exceed 7 cm.

4.4.4. Recommended Governance Measures

The primary goal of the governance measures is to maintain the current bifurcation status, stabilize Biandan Shoal, increase the diversion ratio of the North Baimao Shoal Channel, and control potential increases in the diversion ratio of the ebb tide stream in the North Channel. A comparison of the governance measures for Baimao Shoal revealed that all the proposed measures align with the intended objectives. However, larger variations in the diversion ratio in the South Baimao Shoal Channel would result in a more significant rise in high tidal levels, which could negatively impact flood control and riverbank drainage. As such, two measures with moderate effects are recommended for Baimao Shoal. For Biandan Shoal, three different measures showed similar outcomes. Generally, the higher the crest elevation of the submerged breakwaters is, the greater is the change in the diversion ratio between the North and South Channels, and the larger is the increase in the lowest tidal level in the South Branch. Considering both the cost and environmental impact of the governance measures, the option with the lowest crest elevation for the submerged breakwater is recommended. The combined governance measures demonstrated effects similar to those of the individual measures. After weighing both the positive and negative impacts, the last option in Table 4 was recommended, which has also been adopted in the “Revision of the Comprehensive Renovation and Development Plan for the Changjiang Estuary”. This decision was made to ensure optimal outcomes while taking sustainability and environmental factors into account for the Changjiang Estuary.

It is important to note that the previous analysis indicated that, without protection, the Xingelonggang Channel between the Upper and Lower Biandan Shoals would be subject to scouring. Consequently, the proposed governance measures for Biandan Shoal include a bottom protection project for the Xingelonggang Channel. The project area is shown in Figure 14. Since this project is designed to prevent scouring of the riverbed surface, it will not significantly affect the results and conclusions presented in this study.

5. Conclusions

This paper presents the morphological evolution of the South Branch of the Changjiang Estuary over the past few decades and forecasts its changes over the next 20 years based on both measured data and numerical simulations. The applicability of the numerical model for simulating hydrodynamic and sediment transport processes in the Changjiang Estuary is validated by comparing simulated results with observed data. Based on these simulations, appropriate governance measures are proposed to enhance the stability of the river regime. Several key findings emerge from this study, as summarized below.

Over the past fifty years, the bifurcation configuration of the Changjiang Estuary has remained relatively stable, while the shoals in the South Branch have continuously evolved, with shifts in the diversion ratios. Notably, the diversion ratios of the bifurcated rivers have changed concurrently. This dynamic instability of the shoals and channels presents a significant challenge to the sustainable development of the Changjiang Estuary. The morphological changes in the South Branch between 1958 and 2016 are examined, revealing notable erosion and deposition patterns, with frequent shoal migration and channel incision occurring on decadal timescales. Overall, erosion has dominated the riverbed since 1997. These changes are attributed to multiple factors, including reduced sediment supply and substantial construction activities, such as levees and harbor developments along the banks.

Using a fully validated numerical model, the future morphological evolution of the South Branch is predicted. The results consistently indicate that overall erosion will continue, accompanied by adjustments in the two major channel–shoal systems—Baimao Shoal and Biandan Shoal. Based on these findings, several governance measures to stabilize the system are proposed. Numerical simulations are conducted to assess the feasibility of these measures. The results suggest that increasing the diversion ratio would enhance the stability of the river regime with minimal negative environmental impact, thereby validating the effectiveness of the proposed solutions. It is noteworthy that the governance measures outlined in this study have been incorporated into the new renovation plan for the Changjiang Estuary.